Data-Driven Generative Design (dGD) of a Soft Robot Digital Twin by an Intuitional Evolutionary Algorithm(iEA).

The digital twin concept is pivotal in Industrial 4.0, integrates physical and cyber spaces to address product design challenges. In the early design phase, digital twins offer valuable insights, while topology optimization, often integrated with digital twins, is prevalent during the optimization process. However, traditional topology optimization methods may fall short for soft robots due to their complex physical constraints and the high costs associated with physical simulations. Evolutionary algorithms, with their inherent adaptability, hold promise for autonomous optimization of soft robots based on available data. These algorithms are well-suited for generative design, potentially enabling data-driven generative approaches. This study introduces an intuitional evolutionary algorithm (iEA) to explore a generative optimization method for product design within the context of the digital twin. This paper presents a case study involving the design of soft robots for a speed competition in a simulated 3D environment, demonstrating the viability of combining the iEA with a data-driven generative design (dGD) model to develop a fast- walking soft robot. The results include learning curves and convergence diagrams, illustrating the efficacy of the design solution. The paper will also discuss information hierarchy, requirement analysis and functional modeling of the dGD model. [ABSTRACT FROM AUTHOR]